1. Field of the Invention
The present invention relates to fluid bearings, vacuum chucks, and other devices and methods for producing these items. In one embodiment, the invention relates to a method of manufacturing tools and machinery that may be used during the semiconductor manufacturing process.
2. Description of Related Art
Fluid film bearings are generally formed by a pressurized film of fluid (gas or liquid) contained between two surfaces, conforming to each other with a small gap of approximately uniform thickness existing between the surfaces. These two surfaces may be referred to as the guideway and the fluid film bearing surface or plate. The shape of these members depends on the kind of kinematic constraint realized by the bearing. There are numerous types of fluid bearings, including rotary, cylindrical, flat, spherical, and conical. For example, for rotary motion about an axis, the bearing is formed by two cylindrical, conical or spherical surfaces with a small radial gap between the surfaces of the fluid film bearing plate and the guideway. The fluid film bearing of a spherical pair is free to rotate about the center of the sphere. In one embodiment, the fluid film bearing plate is the moving member and the guideway is the stationary member on which the fluid film bearing plate moves. The converse is also true. The moving member of a cylindrical pair is free to rotate about the axis of the cylinders as well as to translate along the axis.
Typically the bearing is subdivided into several areas, each one having its own bearing surface and restrictors with means for evenly distributing the pressure of the fluid film in order to maximize the load bearing capacity and to achieve optimal bearing stiffness.
Fluid film bearings are formed by either drawing the fluid into the gap by slightly wedging the entrance to the gap and using the fluid viscosity and the motion of the moving member (e.g., fluid bearing) relative to the stationary member (e.g., guideway) to draw the fluid into the gap dynamically, or by externally pressurizing the fluid and pumping it into the gap. This fluid film is delivered to the bearing gap through a pattern or system of grooves (or channels) made in one of the bearing surfaces.
Thus, fluid bearings (and vacuum chucks which are a type of fluid interface) often require a pattern, such as a pattern of grooves, to be created on a surface. An engraving machine, milling machine or stamping press is often used to manufacture a pattern, such as a pattern of grooves on an air bearing or a vacuum chuck. As a result, the patterns are slowly traced along each groove and recreated, each time the pattern is needed, by engraving or milling. This is a time-consuming and costly process. Consequently, very complicated geometries are not often used because of the cost, time and labor involved to mill or engrave such a pattern.
Another method of forming the grooves is by stamping in a stamping press. Stamping the grooves requires using a hard tool containing a protruding pattern of ridges; these ridges, when impressed into an object""s surface make the impression of grooves on the surface of the object. This process deforms the object, extrudes material above the surface which must then be removed, and introduces stresses in the object which must be relieved by a heat treatment process. Moreover, if a complicated geometry is used, it is expensive even for use in mass production.
The bearing gap between the bearing""s surfaces should be uniform, which usually requires that the two surfaces which are separated by the bearing gap conform to each other; that is, the surfaces should xe2x80x9cfitxe2x80x9d to each other as much as possible in much the same way as an idealized finger should fit into an idealized, perfectly matching glove. The pattern of grooves must be engraved, milled or stamped into the bearing surface each time the fluid bearing is made. After the grooves are created, then the surface of the fluid bearing must be lapped or ground to achieve the desired flat, cylindrical, spherical or conical shape. This is required in order to conform the one surface of the fluid bearing to the other surface. If the bearing face of a flat bearing is wavy or otherwise distorted, then the fluid bearing will not adequately support the load that is placed on it and the dynamics of the bearing will be adversely affected. Lapping is a time- and labor consuming and messy process. Because manufacturing fluid bearings is expensive and time-consuming, they are not widely used although they can be beneficial in many machines that require a smooth, straight, controlled motion, such as in positioning stages used in semiconductor equipment or precision machine tools and coordinate-measuring machines.
FIG. 1 illustrates a prior art flat pad air bearing 100 formed by an air bearing body 102 on top of a guideway 116. The combination of the air bearing body 102 and the guideway 116 forms a fluid bearing assembly. The air bearing body 102 is made of a solid block with opening 114 in its side, which provides the air to an air duct hole 110, then to an outlet hole 108 and finally through an orifice 106. A groove 112 is engraved or milled in the face surface 104, which is the surface of the air bearing body 102 that glides along the guideway 116. Typically, the face surface is lapped to obtain a very flat surface which will conform to another flat surface. A front view of a face surface (e.g., 104) is shown as 200 in FIG. 2. Three orifices 202a, 202b, 202c are shown inserted in the face surface 200. A simple pattern of grooves 204a-c has been engraved around each orifice 202a-c. A sill 206 is the area outside the grooves 204. Air escaping out of the grooves 204a-c and past sill 206 builds up pressure, giving the bearing its load bearing capability.
An example of a prior art radially-shaped fluid bearing is shown in FIG. 3A. The view in FIG. 3A of the fluid bearing is from the bearing face surface 300 that glides on a guideway. A cross-section of the fluid bearing of FIG. 3A is shown in FIG. 3B. Four seats must be prepared for the four orifice inserts 303a-303d to rest in the bearing body 309 (shown in FIG. 3B). Each orifice insert 303a-d is coupled to its respective groove 301a-d. Air is provided from the side at 307, typically through a pneumatic fitting (not shown). In FIG. 3B, the orifice may have been too small to drill, so orifice inserts 303b and 303d that have pre-machined smaller orifices (305b and 305d) are used. The smaller orifices 305b and 305d restrict the flow from the air duct 311 to a groove 301b and 301d, respectively. A better design would utilize fewer orifice inserts. But a more efficient and cost-effective design is not practically feasible in the prior art because of the cost, time and labor involved in milling, engraving or stamping grooves in a bearing surface and in lapping the surface.
While prior art techniques for producing fluid bearings or vacuum chucks have used lapping or grinding to achieve conforming surfaces, in an unrelated field, manufacturers of mirrors have used a process whereby a reflective layer is applied to a mirror substrate in such a way that the reflective layer (or layers) conforms to a flatness master. FIG. 3C shows an example of this process which is used to manufacture a mirror. The process 350 shown in FIG. 3C uses a flatness master 351 to which reflective layer 355 and releasing layer 353 are applied. These layers may be applied by known deposition techniques. The flatness master 351 is carefully lapped or ground to be as flat as possible. A mirror substrate 359 is then coated with adhesive, such as an adhesive layer 357 which is flexible before hardening. The mirror substrate 359 is then pressed against the flatness master 351 such that the adhesive layer 357 contacts the layer 355 and hardens while pressed against the layers. The arrow 361 shows the force applied against the mirror substrate 359. The mirror substrate 359 is removed from the flatness master after the adhesive has hardened (cured) enough such that removing the substrate 359 also removes the layers 353 and 355. Now, the reflective layer 355 remains bonded to the mirror substrate 359 and this layer tends to conform to the surface of the flatness master 359. This process of producing a mirror has not been used in the unrelated field of fabrication of fluid bearings or vacuum chucks.
Thus, a time- and cost-effective method of forming patterns, even those with complex geometries, found in certain fluid bearings and vacuum chucks is desirable. Further, a time and cost effective method of forming an optimally conforming surface for fluid interfaces such as fluid bearings or vacuum chucks is also desirable.
The present invention provides fluid bearings and vacuum chucks and methods for forming these devices.
An example of a fluid bearing of the present invention includes a plate support and a flexible bearing plate having a bonding surface which is attached to the plate support with an adhesive which is flexible before hardening. The flexible bearing plate conformed to a predetermined surface which was pressed against the flexible bearing plate during at least a portion of a time that the adhesive hardened.
An example of a method for forming a fluid bearing includes forming a plate having a face surface and a bonding surface opposing the face surface. The plate forms one surface of the fluid bearing. The first side of a body is coupled to the bonding surface of the plate, and the face surface of the plate is placed against a predetermined surface and pressure is used to conform the face surface to the predetermined surface.
One particular embodiment of a method for producing a fluid bearing according to the invention comprises the following steps. First, a bearing plate is formed. The bearing plate has a face surface and a bonding surface opposite the bearing face surface. A pattern may be formed on the face surface of the bearing plate. In one embodiment, the pattern may be etched using standard lithographic techniques. Third, an adhesive is applied to bond a first side of a bearing pad body to the bonding surface of the bearing plate.
Unlike the prior art, the bearing pad body and the bearing plate are not required to have a perfect surface figure at this point in the process. The face surface of the bearing plate is then placed against a flatness master or some other predetermined surface which is designed to conform or model the other surface which opposes the face surface when the fluid bearing is assembled. In one embodiment, the predetermined surface may be the other surface, such as a guideway. A pressure difference between the bonding surface and the face surface of the bearing plate is provided until the face surface replicates or conforms to the shape of the flatness master or the other predetermined surface. The pressure difference may be caused by greater pressure at the bonding surface than at the face surface. For example, the bonding surface can be at atmospheric pressure and the face surface can be at vacuum. Or the bonding surface has a pressure greater than atmospheric pressure and the face surface is at atmosphere or at vacuum. Thus, flatness of the bearing plate and formation of a fluid bearing, in one embodiment, are achieved without the messy and time-consuming process of lapping.
The present invention also teaches methods to form a vacuum chuck or a guideway. Other aspects of the present invention are described below.